parsehub 2.3.10__tar.gz → 2.3.12__tar.gz

{parsehub-2.3.10/src/parsehub.egg-info → parsehub-2.3.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.3.10
+Version: 2.3.12
 Summary: 轻量、异步、开箱即用的社交媒体聚合解析库
 Author-email: 梓澪 <zilingmio@gmail.com>
 License: MIT

{parsehub-2.3.10 → parsehub-2.3.12}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "parsehub"
-version = "2.3.10"
+version = "2.3.12"
 description = "轻量、异步、开箱即用的社交媒体聚合解析库"
 readme = "README.md"
 requires-python = ">=3.12.0"

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/errors.py

@@ -1,11 +1,24 @@
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from .utils.helpers import SecretCookie
 
 
 class ParseHubError(Exception):
     MSG = __doc__
+    cookie: "SecretCookie | None" = None
 
-    def __init__(self, msg: str = ""):
+    def __init__(self, msg: str = "", *, cookie: "SecretCookie | None" = None):
         super().__init__(self.MSG if not msg else f"{self.MSG}: {msg}")
+        self.cookie = cookie
+
+    @property
+    def message(self) -> str:
+        return super().__str__()
+
+    def __str__(self) -> str:
+        return f"{self.message}\n\n使用的 Cookie: {self.cookie}" if self.cookie else self.message
 
 
 class ParseError(ParseHubError):
@@ -34,10 +47,12 @@ def format_attempts(attempts: list[FallbackAttempt], indent: int = 0) -> str:
         if isinstance(attempt.error, FallbackError):
             lines.append(f"{prefix}:")
             lines.append(format_attempts(attempt.error.attempts, indent + 1))
+        elif isinstance(attempt.error, ParseHubError):
+            lines.append(f"{prefix}: {attempt.error.message}")
+            if attempt.error.cookie:
+                lines.append("  " * (indent + 1) + f"使用的 Cookie: {attempt.error.cookie}")
         else:
-            error = attempt.error
-            text = str(error) if isinstance(error, ParseHubError) else f"{type(error).__name__}: {error}"
-            lines.append(f"{prefix}: {text}")
+            lines.append(f"{prefix}: {type(attempt.error).__name__}: {attempt.error}")
     return "\n".join(lines)
 
 

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/base/base.py

@@ -11,6 +11,7 @@ import httpx
 from loguru import logger
 
 from ... import parsers
+from ...errors import ParseHubError
 from ...types import (
     AnyParseResult,
     ContentUnavailableError,
@@ -32,6 +33,16 @@ class FallbackStage(Generic[T]):  # noqa: UP046
     cookie: bool = False
 
 
+def _attach_cookie(e: Exception, cookie: SecretCookie) -> Exception:
+    if isinstance(e, (ContentUnavailableError, FallbackError)):
+        return e
+    if not isinstance(e, ParseHubError):
+        return ParseError(f"{type(e).__name__}: {e}" if str(e) else type(e).__name__, cookie=cookie)
+    if e.cookie is None:
+        e.cookie = cookie
+    return e
+
+
 def is_content_unavailable(e: Exception) -> bool:
     return isinstance(e, ContentUnavailableError)
 
@@ -89,7 +100,13 @@ class BaseParser(ABC):
         """
         self._fallback_attempts = []
         raw_url = await self.get_raw_url(url, clean_all=False)
-        result = await self._do_parse(raw_url)
+        try:
+            result = await self._do_parse(raw_url)
+        except Exception as e:
+            err = _attach_cookie(e, self.cookie)
+            if err is e:
+                raise
+            raise err from e
         result.fallback_errors = list(self._fallback_attempts)
         result.platform = self.__platform__
         raw_url_clean = self._clean_params(raw_url, self.__after_clean_parameters__)
@@ -107,9 +124,13 @@ class BaseParser(ABC):
                 result = await stage.func()
             except Exception as e:
                 logger.opt(exception=e).debug(f"{self.__platform__} {stage.name} 解析失败")
-                if stop(e):
-                    raise
-                attempts.append(FallbackAttempt(stage.name, e, stage.cookie))
+                should_stop = stop(e)
+                err = _attach_cookie(e, self.cookie if stage.cookie else SecretCookie())
+                if should_stop:
+                    if err is e:
+                        raise
+                    raise err from e
+                attempts.append(FallbackAttempt(stage.name, err, stage.cookie))
             else:
                 self._fallback_attempts.extend(attempts)
                 return result

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/parser/bilibili.py

@@ -54,7 +54,7 @@ class BiliParse(BaseParser):
         else:
             return await self._fallback(
                 FallbackStage("API", lambda: self.bili_api_parse(raw_url)),
-                FallbackStage("yt-dlp", lambda: self.ytp_parse(raw_url)),
+                FallbackStage("yt-dlp", lambda: self.ytp_parse(raw_url), cookie=bool(self.cookie)),
                 stop=lambda e: False,
             )
 
@@ -97,7 +97,7 @@ class BiliParse(BaseParser):
             raise
         except Exception as e:
             if "风控" in str(e):
-                raise ParseError(f"账号风控\n使用的cookie: {self.cookie}") from e
+                raise ParseError("账号风控") from e
             raise ParseError(str(e)) from e
         else:
             return cast(BiliDynamic, dynamic_info)

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/parser/instagram.py

@@ -1,9 +1,24 @@
 import re
 
-from ...provider_api.instagram import InstagramAPI, InstagramAPIError, InstagramMediaType, InstagramPost
-from ...types import ImageParseResult, ImageRef, MultimediaParseResult, ParseError, Platform, VideoParseResult, VideoRef
+from ...provider_api.instagram import (
+    InstagramAPI,
+    InstagramAPIError,
+    InstagramMediaType,
+    InstagramPost,
+    InstagramRestrictedError,
+)
+from ...types import (
+    ContentUnavailableError,
+    ImageParseResult,
+    ImageRef,
+    MultimediaParseResult,
+    ParseError,
+    Platform,
+    VideoParseResult,
+    VideoRef,
+)
 from ...utils.helpers import SecretCookie
-from ..base.base import BaseParser
+from ..base.base import BaseParser, FallbackStage
 
 
 class InstagramParser(BaseParser):
@@ -17,7 +32,7 @@ class InstagramParser(BaseParser):
         if not shortcode:
             raise ValueError("Instagram帖子链接无效")
 
-        post = await self._parse(raw_url, shortcode)
+        post = await self._parse(shortcode)
 
         width, height = post.width, post.height
 
@@ -49,25 +64,30 @@ class InstagramParser(BaseParser):
             case _:
                 raise ParseError("不支持的类型")
 
-    async def _parse(self, url: str, shortcode: str, cookie: SecretCookie | None = None) -> InstagramPost:
+    async def _parse(self, shortcode: str) -> InstagramPost:
+        stages: list[FallbackStage[InstagramPost]] = [FallbackStage("匿名", lambda: self._fetch(shortcode))]
+        if self.cookie.get_value():
+            stages.append(
+                FallbackStage("cookie", lambda: self._fetch(shortcode, self.cookie), cookie=True)
+            )
+        return await self._fallback(
+            *stages, stop=lambda e: not isinstance(e.__cause__, InstagramRestrictedError)
+        )
+
+    async def _fetch(self, shortcode: str, cookie: SecretCookie | None = None) -> InstagramPost:
+        api = InstagramAPI(proxy=self.proxy, cookie=cookie.get_value() if cookie else None, timeout=30)
         try:
-            api = InstagramAPI(proxy=self.proxy, cookie=cookie.get_value() if cookie else None, timeout=30)
             return await api.get_post(shortcode)
-        except InstagramAPIError as e:
-            match str(e):
-                case "Fetching Post metadata failed.":
-                    if self.cookie and cookie is None:
-                        return await self._parse(url, shortcode, self.cookie)
-                    else:
-                        raise ParseError("受限视频无法解析: 你必须年满 18 周岁才能观看这个视频") from e
-                case _:
-                    raise ParseError(f"无法获取帖子内容: {e}") from e
+        except ContentUnavailableError:
+            raise
         except Exception as e:
-            if cookie:
-                text = f"Instagram 账号可能已被封禁\n\n使用的Cookie: {cookie}"
-            else:
+            if isinstance(e, InstagramRestrictedError):
                 text = str(e)
-            raise ParseError(f"无法获取帖子内容(可能为私人内容): {text}") from e
+            elif isinstance(e, InstagramAPIError):
+                text = f"无法获取帖子内容: {e}"
+            else:
+                text = f"无法获取帖子内容(可能为私人内容): {e}"
+            raise ParseError(text) from e
 
     @staticmethod
     def get_short_code(url: str) -> str | None:

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/parser/kuaishou.py

@@ -79,8 +79,8 @@ class KuaiShouParser(BaseParser):
                 )
 
             return await self._fallback(
-                FallbackStage("HTML", parse_html),
-                FallbackStage("API", parse_api),
+                FallbackStage("HTML", parse_html, cookie=bool(self.cookie)),
+                FallbackStage("API", parse_api, cookie=bool(self.cookie)),
             )
         else:
             cover = ksp.get_cover_photo_url()

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/parser/twitter.py

@@ -13,12 +13,11 @@ from ...types import (
     AnyMediaRef,
     ImageRef,
     MultimediaParseResult,
-    ParseError,
     Platform,
     RichTextParseResult,
     VideoRef,
 )
-from ..base.base import BaseParser, FallbackStage, is_content_unavailable
+from ..base.base import BaseParser
 
 
 class TwitterParser(BaseParser):
@@ -35,24 +34,7 @@ class TwitterParser(BaseParser):
         return str(urlunparse(urlparse(url)._replace(netloc="x.com")))
 
     async def _parse(self, url: str) -> TwitterTweet:
-        stages: list[FallbackStage[TwitterTweet]] = [
-            FallbackStage("匿名", lambda: Twitter(self.proxy, cookie=None).fetch_tweet(url))
-        ]
-        if cookie := self.cookie.get_value():
-
-            async def fetch_cookie() -> TwitterTweet:
-                try:
-                    return await Twitter(self.proxy, cookie=cookie).fetch_tweet(url)
-                except Exception as e2:
-                    raise ParseError(f"{e2}\n\n使用的 Cookie: {self.cookie}") from e2
-
-            stages.append(FallbackStage("cookie", fetch_cookie, cookie=True))
-
-        return await self._fallback(
-            *stages,
-            stop=lambda e: is_content_unavailable(e)
-            or not any(s in str(e) for s in ("error -2", "error -3")),
-        )
+        return await Twitter(self.proxy, cookie=self.cookie.get_value()).fetch_tweet(url)
 
     @staticmethod
     async def media_parse(tweet: TwitterTweet) -> MultimediaParseResult | RichTextParseResult:

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/parsers/parser/zhihu.py

@@ -14,7 +14,7 @@ from ..base.base import BaseParser
 class ZhihuParser(BaseParser):
     __platform__ = Platform.ZHIHU
     __supported_type__ = ["问答", "专栏", "圈子", "日报"]
-    __match__ = r"^(http(s)?://)?(www|zhuanlan|daily).zhihu.com/(pin|question|p|story)/.*"
+    __match__ = r"^(http(s)?://)?(www|zhuanlan|daily).zhihu.com/(pin|question|p|story|appview/p)/.*"
 
     async def _do_parse(
         self, raw_url: str

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/provider_api/instagram.py

@@ -8,11 +8,17 @@ from typing import Any, cast
 
 import httpx
 
+from ..types import ContentUnavailableError
+
 
 class InstagramAPIError(RuntimeError):
     """Instagram 接口请求或响应解析失败。"""
 
 
+class InstagramRestrictedError(InstagramAPIError):
+    """内容受限, 需要登录 Cookie。"""
+
+
 class InstagramMediaType(StrEnum):
     IMAGE = "GraphImage"
     VIDEO = "GraphVideo"
@@ -129,6 +135,8 @@ class InstagramPost:
 class InstagramAPI:
     GRAPHQL_URL = "https://www.instagram.com/graphql/query"
     INSTAGRAM_URL = "https://www.instagram.com/"
+    RULING_URL = "https://www.instagram.com/api/v1/web/get_ruling_for_content/"
+    APP_ID = "936619743392459"
     SHORTCODE_DOC_ID = "27128499623469141"
 
     DEFAULT_COOKIES = {
@@ -179,11 +187,55 @@ class InstagramAPI:
             },
         )
 
+        if payload.get("data") is None and payload.get("errors"):
+            await self._check_ruling(shortcode)
+
         media = self._extract_shortcode_media(payload)
         if media is None:
-            raise InstagramAPIError("Fetching Post metadata failed.")
+            raise InstagramRestrictedError("受限视频无法解析: 你必须年满 18 周岁才能观看这个视频")
         return media
 
+    async def _check_ruling(self, shortcode: str) -> None:
+        """查询内容裁决接口, 在 GraphQL 报错时解释原因。失败时静默返回。"""
+        try:
+            async with self._new_client() as client:
+                try:
+                    await self._ensure_csrf_token(client)
+                except InstagramAPIError:
+                    return
+                response = await client.get(
+                    self.RULING_URL,
+                    params={
+                        "content_type": "MEDIA",
+                        "target_id": str(self.shortcode_to_media_id(shortcode)),
+                    },
+                    headers={
+                        "X-IG-App-ID": self.APP_ID,
+                        "Accept-Language": "zh-CN,zh;q=0.9",
+                    },
+                )
+            body = response.json()
+        except (httpx.HTTPError, ValueError):
+            return
+        if not isinstance(body, dict):
+            return
+
+        if body.get("message") == "Media cannot be found":
+            raise ContentUnavailableError("帖子不存在或已被删除")
+        if title := body.get("title"):
+            description = body.get("description")
+            raise InstagramRestrictedError(f"{title}: {description}" if description else title)
+
+    @staticmethod
+    def shortcode_to_media_id(shortcode: str) -> int:
+        if len(shortcode) > 28:
+            shortcode = shortcode[:-28]
+        alphabet = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_"
+        media_id = 0
+        for char in shortcode:
+            media_id = media_id * 64 + alphabet.index(char)
+        return media_id
+
     async def _post_graphql(self, *, doc_id: str, variables: dict[str, Any]) -> dict[str, Any]:
         async with self._new_client() as client:
             await self._ensure_csrf_token(client)
@@ -376,5 +428,6 @@ __all__ = [
     "InstagramAPIError",
     "InstagramMediaType",
     "InstagramPost",
+    "InstagramRestrictedError",
     "InstagramSidecarNode",
 ]

{parsehub-2.3.10 → parsehub-2.3.12}/src/parsehub/provider_api/zhihu.py

@@ -211,7 +211,7 @@ class ZhihuAPI:
     async def parse(self, raw_url: str) -> ZhihuQA | ZhihuZhuanLan | ZhihuPin:
         if "/question" in raw_url:
             return await self.parse_qa(raw_url)
-        if "zhuanlan." in raw_url:
+        if "zhuanlan." in raw_url or "/appview/p/" in raw_url:
             return await self.parse_zl(raw_url)
         if "/pin/" in raw_url:
             return await self.parse_pin(raw_url)

{parsehub-2.3.10 → parsehub-2.3.12/src/parsehub.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: parsehub
-Version: 2.3.10
+Version: 2.3.12
 Summary: 轻量、异步、开箱即用的社交媒体聚合解析库
 Author-email: 梓澪 <zilingmio@gmail.com>
 License: MIT